CN106646096B - Transformer fault classification and recognition methods based on vibration analysis method - Google Patents
Transformer fault classification and recognition methods based on vibration analysis method Download PDFInfo
- Publication number
- CN106646096B CN106646096B CN201611003127.4A CN201611003127A CN106646096B CN 106646096 B CN106646096 B CN 106646096B CN 201611003127 A CN201611003127 A CN 201611003127A CN 106646096 B CN106646096 B CN 106646096B
- Authority
- CN
- China
- Prior art keywords
- sample
- transformer
- intrinsic mode
- mode functions
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000004458 analytical method Methods 0.000 title claims abstract description 17
- 238000012360 testing method Methods 0.000 claims abstract description 37
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 17
- 238000000513 principal component analysis Methods 0.000 claims abstract description 9
- 238000003909 pattern recognition Methods 0.000 claims abstract description 8
- 230000009467 reduction Effects 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 26
- 238000012549 training Methods 0.000 claims description 16
- 238000004804 winding Methods 0.000 claims description 15
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims description 2
- 230000004069 differentiation Effects 0.000 claims 1
- 239000000284 extract Substances 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000002405 diagnostic procedure Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/72—Testing of electric windings
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a kind of transformer fault classification and recognition methods based on vibration analysis method, comprising the following steps: S1: choosing transformer test object, and acquiring transformer vibration signal under different conditions is sample data;S2: intrinsic mode functions are obtained using ensemble empirical mode decomposition method calculating sample data is gathered in Hilbert-Huang transform;S3: characteristic vector V in intrinsic mode functions component is extracted;S4: dimensionality reduction is carried out to characteristic vector using Principal Component Analysis, coordinate projection is into two dimensional image;S5: classified adjacent to method to sample data using K;S6: test sample is calculated at a distance from original sample using range formula;S7: pattern-recognition is carried out;S8: corresponding transformer fault type in output mode identification;It can intuitively and effectively judge running state of transformer.
Description
Technical field
The present invention relates to electric device maintenance fields, and in particular to a kind of transformer fault classification based on vibration analysis method
And recognition methods.
Background technique
In the various equipment of electric system, transformer is expensive and important equipment, and reliability is for guaranteeing power grid
It is safely operated significant.Transformer fault statistics over the years show transformer winding and iron core is the more portion that breaks down
The Accident of Transformer that part, winding and iron core cause proportion in the total accident of transformer reaches 70%.
Currently, the method for detection running state of transformer is increasing.Short circuit impedance method passes through the electricity of off-line measurement winding
Resist and observe its change in impedance value to judge winding deformation situation, but this method sensitivity is low, poor reliability.Low Voltage Impulse Method,
It is mainly used in field test test, however this method has vulnerable to interference effect, need to correct, tests the special cloth of circuit requirement
The defects of setting the repeatability to guarantee test.Frequency Response Analysis method is carried out by the frequency response curve of 3 phase winding of measuring transformer
Test transformer winding deformation degree overcomes some defects that Low Voltage Impulse Method has repeatability difference, but this method can only
Offline inspection is carried out to transformer.But three of the above method can only all realize offline inspection, it therefore, can be with there is an urgent need to one kind
The method for carrying out carrying out live detection to transformer fault.
Existing set ensemble empirical mode decomposition method, i.e. EEMD method extract feature vector and the utilization of transformer state
Fisher diagnostic method, which carries out failure modes, may be implemented the live detection to transformer fault.But transformation is extracted according to EEMD
The characteristic vector of device vibration signal, characteristic vector are multi-group datas, and when test sample is more, and test sample can become one
Multidimensional numerical.Multidimensional numerical is unfavorable for calculating, while the observation being also unfavorable in reality;Fisher diagnostic method can only be realized simultaneously
Linear classification cannot achieve Nonlinear Classification, due to the uncertainty of test measurement data in test data, inevitably will appear
Wrong data, these wrong data influence whether the effect of linear classification.
Summary of the invention
In order to solve the above-mentioned technical problem the present invention provides a kind of transformer fault classification and knowledge based on vibration analysis method
Other method.
The present invention is achieved through the following technical solutions:
Transformer fault classification and recognition methods based on vibration analysis method, comprising the following steps:
S1: choosing transformer test object, and acquiring transformer vibration signal under different conditions is sample data;
S2: eigen mode letter is obtained using ensemble empirical mode decomposition method calculating sample data is gathered in Hilbert-Huang transform
Number;
S3: characteristic vector V in intrinsic mode functions component is extracted;
S4: dimensionality reduction is carried out to characteristic vector using Principal Component Analysis, coordinate projection is into two dimensional image;
S5: classified adjacent to method to sample data using K;
S6: test sample is calculated at a distance from original sample using range formula;
S7: pattern-recognition is carried out;
S8: corresponding transformer fault type in output mode identification.
Due to the presence of transformer winding and core structure nonlinear characteristic, transformer vibration signal can show brighter
Aobvious is non-linear.Traditional Time-Frequency Analysis Method is in analysis nonlinear and non local boundary value problem Shortcomings.The present invention uses principal component
Analytic approach and KNN Classification and Identification method realize the intuitive taxonomy to test sample and pattern-recognition.It, can for test sample data
To realize intuitive projection on 2d, reach intuitive taxonomy effect, the minute quantity wrong data in classification will not shadow
Ring the effect to KNN classification and pattern-recognition.
In step s 2, sample data is calculated using set ensemble empirical mode decomposition method in Hilbert-Huang transform to obtain
The step of intrinsic mode functions are as follows:
To the Gaussian white noise sequence n that M times is added in original signal x (t)i(t), (i=1,2 ..., M), it may be assumed that
Xi(t)=x (t)+ni(t);
To Xi(t) empirical mode decomposition is carried out respectively, obtains the component and surplus of each intrinsic mode functions, it may be assumed that
Wherein hijAfter white Gaussian noise is added for i-th, to Xi(t) j-th of the intrinsic mode functions decomposed point
Amount;rinAfter white Gaussian noise is added for i-th, to Xi(t) remainder after being decomposed;N is Decomposition order;
Using the zero-mean principle of white Gaussian noise frequency spectrum, the corresponding intrinsic mode functions component of above step is carried out overall
Average calculating operation, eliminate white Gaussian noise influences as time domain distribution reference structure bring, intrinsic after finally obtained EEMD
The component of modular function are as follows:
In formula, hj(t) it indicates to carry out j-th of intrinsic mode functions component that EEMD is decomposed to original signal;M is to be added
The number of white noise.Wherein, intrinsic mode functions, that is, IMF;Gathering ensemble empirical mode decomposition method in Hilbert-Huang transform is
EEMD;Empirical mode decomposition, that is, EMD.
In step s3, characteristic vector V calculation method in intrinsic mode functions component is extracted, each IMF component is chosen and carries out
Hilbert analysis, and to the amplitude constitutive characteristic vector V under corresponding:
V=[v1, v2..., vz];
In formula, Z indicates characteristic vector number, AjIt (i) is the amplitude of j-th of intrinsic mode functions component.
In step s 4, steps are as follows to characteristic vector progress dimensionality reduction for Principal Component Analysis:
Using EEMD extract transformer vibration signal under 3 kinds of normal condition, winding failure, iron core failure operating conditions feature to
Amount
V=[v1, v2..., vz];
It is directed to 3 kinds of operating conditions simultaneously, it is each to select m group empirically sample, obtain the training matrix R of 4m × z;
Centralization is done to training matrix R to handle to obtain matrix A=[aij],
Wherein, aijIt is the data in R after element center, vijIt is the sample value that the i-th row jth arranges in training matrix R,It is
The mean value of every row in training matrix R;
Using calculate centralization matrix A covariance matrix S,
The characteristic value and feature vector of covariance matrix S in calculating carries out maximum value arrangement, takes the first two nonzero eigenvalue
Corresponding feature vector [a1, a2] it is projecting direction;
To [a shown in R progress1, a2] direction projection transform, obtain the X-Y scheme of training sample:
In formula, T representing matrix transposition, Y1、Y2The respectively cross, ordinate of two-dimensional projection.
Principal component analytical method, that is, PCA is the position distribution according to sample point in multi-dimensional model space, most generous with variance
To the feature extraction and data compression of data are realized as vector is differentiated in the i.e. maximum direction that changes in space of sample point.
In step s 6, it calculates test sample and the range formula of original sample is as follows:
Test sample vector Yt=[yt1, yt2]。
In the step s 7, mode identification method is as follows:
The K neighbour apart from K relatively small sample as test sample is selected, if w is transformer state type, K is a
In group sample, wherein come from w1The sample of Status Type has M1It is a, come from w2The sample of Status Type has M2It is a ..., come from wc
The sample of Status Type has McIt is a, if k1, k2..., kcIt is to belong to w in k neighbour respectively1, w2..., wcThe sample number of class, then it is fixed
Justice differentiates letter are as follows:
gi(Vt)=ki, i=1,2 ..., c;
If gj(Vt)=max (kt), then test sample Vi∈ωj, the corresponding variable-pressure operation state of test sample x is wj transformation
Device Status Type.
K-nearest neighbor, that is, K-Nearest Neighbor, KNN is to calculate sample x to be sorted and instruction according to distance metric function
Practice the distance for concentrating each training sample, to calculated distance-taxis, selection and K nearest training sample of sample to be sorted
The K arest neighbors as x.
The EEMD method that Huang is proposed has the characteristic of zero-mean using white noise, and EEMD method is in signal to be decomposed
Empirical mode decomposition is carried out after white noise is added, after multiple averaging, noise will be supported mutually decomposition result, integrate the knot of mean value
Fruit is as final IMF component.A large amount of transformer winding vibration signal datas are tested, the intrinsic mode functions that EEMD is obtained
Intrinsic Mode IMF can embody the vibration characteristics of transformer winding well, and have specific physical significance.Therefore,
Transformer vibration signal, which is carried out EEMD selection, can effectively reflect the IMF of transformer vibration information, calculate its instantaneous frequency and wink
Between amplitude as characteristic vector, can be with quantificational expression transformer current state.
Compared with prior art, the present invention at least having the following advantages and benefits:
The present invention is analyzed using vibration signal row of the EEMD to transformer, extracts Transformer nominal situation, winding
Vibration signal extracts characteristic vector under deformation, iron core failure different conditions, and vibration signal is extracted using principal component analysis
Characteristic vector carries out Dimension Reduction Analysis, and characteristic vector is projected in intuitive X-Y scheme, finally uses KNN classifying identification method
Classification and pattern-recognition are carried out to data, to intuitively and effectively judge running state of transformer.
Detailed description of the invention
Attached drawing described herein is used to provide to further understand the embodiment of the present invention, constitutes one of the application
Point, do not constitute the restriction to the embodiment of the present invention.In the accompanying drawings:
Fig. 1 is the flow chart of the method for the present invention.
Fig. 2 is the sample data result figure in the embodiment of the present invention 2.
Fig. 3 is the test sample pattern recognition result figure in the embodiment of the present invention 2.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below with reference to embodiment and attached drawing, to this
Invention is described in further detail, and exemplary embodiment of the invention and its explanation for explaining only the invention, are not made
For limitation of the invention.
Embodiment 1
Transformer fault classification and recognition methods based on vibration analysis method as shown in Figure 1, comprising the following steps:
S1: choosing corresponding transformer is subjects, and acquiring transformer vibration signal under different conditions is sample data;
S2: eigen mode letter is obtained using ensemble empirical mode decomposition method calculating sample data is gathered in Hilbert-Huang transform
Number;
S3: characteristic vector V in intrinsic mode functions component is extracted;
S4: dimensionality reduction is carried out to characteristic vector using Principal Component Analysis, coordinate projection is into two dimensional image;
S5: classified adjacent to method to sample data using K;
S6: test sample is calculated at a distance from original sample using range formula;
S7: pattern-recognition is carried out;
S8: corresponding transformer fault type in output mode identification.
Embodiment 2,
On the basis of the above embodiments, specific:
In step s 2, sample data is calculated using set ensemble empirical mode decomposition method in Hilbert-Huang transform to obtain
The step of intrinsic mode functions are as follows:
To the Gaussian white noise sequence n that M times is added in original signal x (t)i(t), (i=1,2 ..., M), it may be assumed that
Xi(t)=x (t)+ni(t);
To Xi(t) EMD is carried out respectively, obtains the component and surplus of each intrinsic mode functions, it may be assumed that
Wherein hijAfter white Gaussian noise is added for i-th, to Xi(t) j-th of the intrinsic mode functions decomposed point
Amount;rinAfter white Gaussian noise is added for i-th, to Xi(t) remainder after being decomposed;N is Decomposition order;
Using the zero-mean principle of white Gaussian noise frequency spectrum, the corresponding intrinsic mode functions component of above step is carried out overall
Average calculating operation, eliminate white Gaussian noise influences as time domain distribution reference structure bring, intrinsic after finally obtained EEMD
The component of modular function are as follows:
In formula, hj(t) it indicates to carry out j-th of intrinsic mode functions component that EEMD is decomposed to original signal;N is to be added
The number of white noise.
In step s3, characteristic vector V calculation formula in intrinsic mode functions component: V=[v is extracted1, v2..., vz];
In formula, AjIt (i) is the amplitude of j-th of intrinsic mode functions component.
In step s 4, steps are as follows to characteristic vector progress dimensionality reduction for Principal Component Analysis:
Using EEMD extract transformer vibration signal under 3 kinds of normal condition, winding failure, iron core failure operating conditions feature to
Amount
V=[v1, v2..., vz];
It is directed to 3 kinds of operating conditions simultaneously, it is each to select m group empirically sample, obtain the training matrix R of 4m × z;
Centralization is done to training matrix R to handle to obtain matrix A=[aij],
Wherein, aijIt is the data in R after element center, vijIt is the sample value that the i-th row jth arranges in training matrix R,It is
The mean value of every row in training matrix R;
Using calculate centralization matrix A covariance matrix S,
The characteristic value and feature vector of covariance matrix S in calculating carries out maximum value arrangement, takes the first two nonzero eigenvalue
Corresponding feature vector [a1, a2] it is projecting direction;
To [a shown in R progress1, a2] direction projection transform, obtain the X-Y scheme of training sample;
In formula, Y1、Y2The respectively cross, ordinate of two-dimensional projection.
In step s 6, it calculates test sample and the range formula of original sample is as follows:
Test sample vector Yt=[yt1, yt2]。
In the step s 7, mode identification method is as follows:
The K neighbour apart from K relatively small sample as test sample is selected, if w is transformer state type, K is a
In group sample, w is come from1The sample of Status Type has M1It is a, come from w2The sample of Status Type has M2It is a ..., come from wcState class
The sample of type has McIt is a to come from, if k1, k2..., kcIt is to belong to w in k neighbour respectively1, w2..., wcThe sample number of class, then define
Differentiate letter are as follows:
gi(Vt)=ki, i=1,2 ..., c;
If gj(Vt)=max (kt), then test sample Vi∈ωj, the corresponding variable-pressure operation state of test sample x is wjTransformation
Device Status Type.
To verify the above-mentioned transformer fault detection method based on principal component analysis and KNN classifying identification method, this test
It is analyzed and researched using the vibration data that a 110kv three-phase transformer of Sichuan Guangan Utilities Electric Co. obtains.In experiment,
By 3 ICP type acceleration vibrating sensors, sensitivity 100mv/g is individually positioned in transformer height in a manner of permanent magnet
Press on side box wall, in the middle part of the corresponding tank wall of every phase winding, bottom and wall sides position carry out sampled data.Vibration data is adopted
Sample frequency is 25.6kHz.Simulation result is as shown in Figures 2 and 3.It is proved by test, this method may be implemented to transformer just
Normal state, winding deformation, iron core failure three state intuitive taxonomy, and quick automatic mode identification is carried out to test sample.This
Invention emulation case is only intended to help to illustrate the present invention.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not intended to limit the present invention the foregoing is merely a specific embodiment of the invention
Protection scope, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (4)
1. transformer fault classification and recognition methods based on vibration analysis method, which comprises the following steps:
S1: choosing transformer test object, and acquiring transformer vibration signal under different conditions is sample data;
S2: intrinsic mode functions are obtained using ensemble empirical mode decomposition method calculating sample data is gathered in Hilbert-Huang transform;
S3: characteristic vector V in intrinsic mode functions component is extracted;
S4: dimensionality reduction is carried out to characteristic vector using Principal Component Analysis, coordinate projection is into two dimensional image;
S5: classified adjacent to method to sample data using K;
S6: test sample is calculated at a distance from original sample using range formula;
S7: pattern-recognition is carried out;
S8: corresponding transformer fault type in output mode identification;
In the step S4, Principal Component Analysis carries out dimensionality reduction to characteristic vector, and steps are as follows:
Transformer vibration signal feature vector V under 3 kinds of normal condition, winding failure, iron core failure operating conditions is extracted using EEMD
=[v1, v2..., vz];
It is directed to 3 kinds of operating conditions simultaneously, it is each to select m group empirically sample, obtain the training matrix R of 4m × z;
Centralization is done to training matrix R to handle to obtain matrix A=[aij],
Wherein, aijIt is the data in R after element center, vijIt is the sample value that the i-th row jth arranges in training matrix R, viIt is trained
The mean value of every row in matrix R;
Using calculate centralization matrix A covariance matrix S,
The characteristic value and feature vector of covariance matrix S in calculating carries out maximum value arrangement, takes the first two nonzero eigenvalue corresponding
Feature vector [α1, α2] it is projecting direction;
To [α shown in R progress1, α2] direction projection transform, obtain the X-Y scheme of training sample;
In formula, T representing matrix transposition, Y1、Y2The respectively cross, ordinate of two-dimensional projection;
In the step S7, mode identification method is as follows:
K neighbour of the K sample as test sample is selected, if w is transformer state type, in K group sample, from w1Shape
The sample of state type has M1It is a, come from w2The sample of Status Type has M2It is a ..., come from wcThe sample of Status Type has McIt is a, if
k1, k2..., kcIt is to belong to w in k neighbour respectively1, w2..., wcThe sample number of class then defines differentiation letter are as follows:
gi(Vt)=ki, i=1,2 ..., c;
If gj(Vt)=max (ki), then test sample Vi∈ωj, the corresponding variable-pressure operation state of test sample x is wjTransformer shape
State type.
2. the transformer fault classification and recognition methods according to claim 1 based on vibration analysis method, it is characterised in that:
In step s 2, sample data is calculated using set ensemble empirical mode decomposition method in Hilbert-Huang transform obtain eigen mode letter
Several steps are as follows:
To the Gaussian white noise sequence n that M times is added in original signal x (t)i(t), (i=1,2 ..., M) obtains Xi(t), it may be assumed that
Xi(t)=x (t)+ni(t);
To Xi(t) empirical mode decomposition is carried out respectively, obtains the component and surplus of each intrinsic mode functions, it may be assumed that
Wherein hijAfter white Gaussian noise is added for i-th, to Xi(t) j-th of the intrinsic mode functions component decomposed;rin
After white Gaussian noise is added for i-th, to Xi(t) remainder after being decomposed;N is Decomposition order;
Using the zero-mean principle of white Gaussian noise frequency spectrum, the corresponding intrinsic mode functions component of above step is subjected to population mean
Operation, the component of the intrinsic mode functions after obtained EEMD are as follows:
In formula, hj(t) it indicates to carry out j-th of intrinsic mode functions component that EEMD is decomposed to original signal;M is that white noise is added
The number of sound.
3. the transformer fault classification and recognition methods according to claim 1 based on vibration analysis method, it is characterised in that:
In step s3, characteristic vector V calculation formula in intrinsic mode functions component: V=[v is extracted1, v2..., vz];
In formula, Z indicates characteristic vector number;AjIt (i) is the amplitude of j-th of intrinsic mode functions component.
4. the transformer fault classification and recognition methods according to claim 1 based on vibration analysis method, it is characterised in that:
In step s 6, it calculates test sample and the range formula of original sample is as follows:
Test sample vector Yt=[yt1, yt2]。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611003127.4A CN106646096B (en) | 2016-11-15 | 2016-11-15 | Transformer fault classification and recognition methods based on vibration analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611003127.4A CN106646096B (en) | 2016-11-15 | 2016-11-15 | Transformer fault classification and recognition methods based on vibration analysis method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106646096A CN106646096A (en) | 2017-05-10 |
CN106646096B true CN106646096B (en) | 2019-07-02 |
Family
ID=58806585
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611003127.4A Active CN106646096B (en) | 2016-11-15 | 2016-11-15 | Transformer fault classification and recognition methods based on vibration analysis method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106646096B (en) |
Families Citing this family (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107037298B (en) * | 2017-06-21 | 2019-12-10 | 广州供电局有限公司 | Transformer direct current bias detection method and device, storage medium and computer equipment |
CN107356427A (en) * | 2017-07-10 | 2017-11-17 | 成都航空职业技术学院 | A kind of electric machine iron core method for detecting abnormality based on empirical mode decomposition |
CN108280467B (en) * | 2018-01-12 | 2020-10-16 | 北京摩拜科技有限公司 | Vehicle fault detection method, detection equipment, server and vehicle system |
CN108510002B (en) * | 2018-04-02 | 2020-02-21 | 西南交通大学 | Method for detecting short circuit impact resistance of wound core traction transformer winding |
CN108872863B (en) * | 2018-05-02 | 2020-09-08 | 广东工业大学 | Optimized and classified electric vehicle charging state monitoring method |
CN108983042B (en) * | 2018-07-25 | 2020-05-22 | 国网湖北省电力有限公司电力科学研究院 | KNN-based distribution network ground fault reason identification method |
CN109374323A (en) * | 2018-09-29 | 2019-02-22 | 国网山西省电力公司阳泉供电公司 | Transformer mechanical fault detection method based on vibration signal index energy |
CN109374119A (en) * | 2018-09-29 | 2019-02-22 | 国网山西省电力公司阳泉供电公司 | Transformer vibration signal Characteristic Extraction method |
CN109813417A (en) * | 2019-01-18 | 2019-05-28 | 国网江苏省电力有限公司检修分公司 | A kind of shunt reactor method for diagnosing faults based on improvement EMD |
CN109901030B (en) * | 2019-02-25 | 2021-10-22 | 国网山东省电力公司济南供电公司 | Reactor turn-to-turn insulation state monitoring method, system and application |
CN110160781B (en) * | 2019-06-18 | 2020-12-04 | 广东石油化工学院 | Test set reconstruction and prediction method for rotary machine fault classification |
CN110348534A (en) * | 2019-07-12 | 2019-10-18 | 国网四川省电力公司信息通信公司 | Data Dimensionality Reduction and characteristic analysis method in a kind of flow analysis |
US11585692B2 (en) * | 2019-10-24 | 2023-02-21 | Palo Alto Research Center Incorporated | Fiber optic sensing system for grid-based assets |
US11719559B2 (en) | 2019-10-24 | 2023-08-08 | Palo Alto Research Center Incorporated | Fiber optic sensing system for grid-based assets |
CN111239521B (en) * | 2020-02-18 | 2022-02-11 | 金陵科技学院 | Wind power generation system converter fault identification method based on PCA-kNN |
CN112070999B (en) * | 2020-08-11 | 2022-05-31 | 上海工程技术大学 | Underground pipe network protection warning board and vibration signal identification method thereof |
CN112052890B (en) * | 2020-08-28 | 2024-04-02 | 华北电力科学研究院有限责任公司 | Method and device for predicting vibration of water supply pump |
CN112711865B (en) * | 2021-01-19 | 2022-09-02 | 国网河北省电力有限公司 | Power transmission line inductive electric simulation refined modeling method |
CN113191437A (en) * | 2021-05-07 | 2021-07-30 | 国网山西省电力公司电力科学研究院 | Transformer mechanical fault detection method based on vibration signal composite eigenvector |
CN113989201B (en) * | 2021-10-15 | 2023-04-07 | 中国大唐集团科学技术研究院有限公司火力发电技术研究院 | Axle center track identification method based on cloud computing and LSTM |
CN114966507B (en) * | 2022-05-27 | 2022-12-13 | 浙江大学 | Method for evaluating intrinsic time domain stability of radio frequency receiving coil in fMRI |
CN116465623B (en) * | 2023-05-10 | 2023-09-19 | 安徽大学 | Gearbox service life prediction method based on sparse converter |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010060253A1 (en) * | 2008-11-25 | 2010-06-03 | 上海市电力公司 | The system and method for detecting the state of the transformer winding by utilizing excitation of the constant current sweep frequency power source |
CN101750210A (en) * | 2009-12-24 | 2010-06-23 | 重庆大学 | Fault diagnosis method based on OLPP feature reduction |
CN103258134A (en) * | 2013-05-14 | 2013-08-21 | 宁波大学 | Dimension reduction processing method of high-dimensional vibration signals |
CN103295025A (en) * | 2013-05-03 | 2013-09-11 | 南京大学 | Automatic selecting method of three-dimensional model optimal view |
CN103499382A (en) * | 2013-09-30 | 2014-01-08 | 国家电网公司 | Vibration data fusion and image recognition-based transformer fault diagnosis method |
CN105547465A (en) * | 2015-12-08 | 2016-05-04 | 华北电力大学(保定) | Transformer vibration signal winding state feature extraction method |
CN105652120A (en) * | 2015-12-31 | 2016-06-08 | 国网重庆潼南区供电有限责任公司 | Power transformer fault detection method and detection system |
-
2016
- 2016-11-15 CN CN201611003127.4A patent/CN106646096B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010060253A1 (en) * | 2008-11-25 | 2010-06-03 | 上海市电力公司 | The system and method for detecting the state of the transformer winding by utilizing excitation of the constant current sweep frequency power source |
CN101750210A (en) * | 2009-12-24 | 2010-06-23 | 重庆大学 | Fault diagnosis method based on OLPP feature reduction |
CN103295025A (en) * | 2013-05-03 | 2013-09-11 | 南京大学 | Automatic selecting method of three-dimensional model optimal view |
CN103258134A (en) * | 2013-05-14 | 2013-08-21 | 宁波大学 | Dimension reduction processing method of high-dimensional vibration signals |
CN103499382A (en) * | 2013-09-30 | 2014-01-08 | 国家电网公司 | Vibration data fusion and image recognition-based transformer fault diagnosis method |
CN105547465A (en) * | 2015-12-08 | 2016-05-04 | 华北电力大学(保定) | Transformer vibration signal winding state feature extraction method |
CN105652120A (en) * | 2015-12-31 | 2016-06-08 | 国网重庆潼南区供电有限责任公司 | Power transformer fault detection method and detection system |
Also Published As
Publication number | Publication date |
---|---|
CN106646096A (en) | 2017-05-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106646096B (en) | Transformer fault classification and recognition methods based on vibration analysis method | |
CN106443310B (en) | A kind of transformer fault detection method based on SOM neural network | |
Wenrong et al. | Digital detection, grouping and classification of partial discharge signals at DC voltage | |
CN108573225B (en) | Partial discharge signal pattern recognition method and system | |
JP6476861B2 (en) | Electromagnetic field feature classification presentation device | |
CN108957251A (en) | A kind of cable connector Partial Discharge Pattern Recognition Method | |
CN106443316A (en) | Power transformer winding deformation state multi-information detection method and device | |
CN109033612A (en) | A kind of Diagnosis Method of Transformer Faults based on vibration noise and BP neural network | |
CN105404867B (en) | A kind of substation isolating-switch state identification method of view-based access control model | |
CN109165604A (en) | The recognition methods of non-intrusion type load and its test macro based on coorinated training | |
CN104849633A (en) | Switchgear partial discharge mode recognition method | |
CN109375010A (en) | Winding in Power Transformer method for diagnosing faults based on vibration distribution characteristic model | |
CN103116090A (en) | Three-phrase pulse-width modulation (PWM) rectifier fault diagnosis method based on wavelet packet analysis and support vector machine | |
Wang et al. | Few-shot transfer learning with attention mechanism for high-voltage circuit breaker fault diagnosis | |
CN110070102B (en) | Method for establishing sequence-to-sequence model for identifying power quality disturbance type | |
CN106548013B (en) | Utilize the voltage sag source identification method for improving incomplete S-transformation | |
CN104778348B (en) | Voltage sag action characteristic modeling method for low-voltage trip | |
CN106707094A (en) | Classification recognition method of arc fault of low voltage power supply and distribution line | |
CN104713714B (en) | A kind of primary cut-out action characteristic analysis method clustered based on grid multi-density | |
CN110175571A (en) | The intellectual monitoring of substation equipment state and recognition methods | |
CN110082082A (en) | A kind of GIS state identification method based on vibration signal Principal Component Analysis | |
CN109828168A (en) | Converter method for diagnosing faults based on Density Estimator | |
CN113884844A (en) | Transformer partial discharge type identification method and system | |
CN111999591A (en) | Method for identifying abnormal state of primary equipment of power distribution network | |
CN117110798B (en) | Fault detection method and system for intelligent power distribution network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |